back to
EGAD homepage
Table of
Contents
1. Introduction
2. Descriptions of main, program flow, overview of data-structures
3. Inputfiles
4. pdb file input, modifiers, and related jobs
5. Energy function files and modifiers
6. JOBTYPE point_energy and JOBTYPE
sasa_calc
7. JOBTYPE energy_profile_table
8. Input of
variable and fixed positions for rotamer optimization
9. Pair energy lookup table generation
10. Rotamer optimization jobs
11. Prediction of pKa's - JOBTYPE pK
12. Ligands
13. Parallelization of EGAD jobs
14. Scanning mutagenesis
15. Batch jobs
16. Protein-protein complexes
17. Sequence restraints on design
18. Analyses and optimizations that require
consideration of multiple structures/states
19. Building the backbone with
ideal geometry
20. Dihedral
angle minimization - JOBTYPE minimize
Appendices:
A.1 Output pdb file format
A.2 Energy function file formats
References
Examples directory
Source code directory
download EGAD User's Manual (pdf)
download
Navin Pokala's thesis (pdf)
download the EGAD
program
with example files
installation instructions
1. Introduction
1.1 Overview
1.2 Installation
1.2.1
Compilation and execution issues
1.3 EGAD source code
2. Descriptions of main,
program flow, overview of data-structures
2.1 main
2.2 Exit
status, errors, I/O errors,
and warnings
2.3 Premature exit or early
program advancement
2.4 Data structure overview
3. Inputfiles
3.1 General
layout and rules
3.2 Sequence position strings
for input and output
3.3 Comments
3.4 OUTPUT_PREFIX
3.5 Description
of the input_stuff.cpp:
input_stuff function
4. pdb file input,
modifiers, and related jobs
4.1
Requirements for template pdb
files, special cases, and modifiers
4.1.1
Missing atoms
4.1.2
User-defined rotamers
4.1.3 TRANSFORM_MATRIX
4.1.4 Rebuilding
structures from extracted dihedral angles (REBUILD_FLAG)
4.1.5 Flipping of
Asn, Gln, and His rotamers
4.1.6
Sequence numbering and atom
order
4.1.7
Chain IDs and SUPER_CHAIN
4.1.8
Disulfide bonds
4.1.9
Termini
4.1.10
Ligands
4.2 JOBTYPE
HBUILD
4.3 JOBTYPE
ALIGN_STRUCTURES
Figure 4.3.1 - Overlay of bound
and free maltose-binding protein reveals a hinge motion
4.4
Description of the readpdbfile.cpp:
readpdbfile function
5. Energy function
files and modifiers
5.1 Overview
Figure
5.1.1 - Energy spectrum for protein design
5.2 Energy function modifiers
5.3 Description of energy
function datastructures
and input functions
5.3.1
Overview of energy function datastructures
5.3.2
Description of read_forcefield.cpp: read_forcefield
6. JOBTYPE
point_energy and JOBTYPE sasa_calc
6.1
Overview
6.2 Implementation details
7. JOBTYPE
energy_profile_table
8. Input of variable
and fixed positions for
rotamer optimization
8.1 Overview of
rotamer optimization
8.2 Explicit definition of
variable and fixed
positions
8.2.1
Permitted
residuetype string formats
8.2.2
Proline, glycine, and
disulfide-bonded cysteine positions are not allowed to vary
8.2.3 OTHER_POSITIONS modifier
8.2.4 AVOID_NATIVE_ROTAMER_FLAG modifier
8.3 Shortcuts for
non-explicit definition
of variable positions
8.3.1
Input of a SEQUENCE
8.3.2 JOBTYPE list_variable_positions
8.4
Description of variable position input structs and functions
8.4.1 VARIABLE_POSITION
and CHOICE datastructures
8.4.2 input_stuff.cpp:
input_VARIABLE_POSITION
8.4.3 VARIABLE_POSITIONS.cpp:
modify_VARIABLE_POSITION
8.4.4 input_stuff.cpp:
sequence_to_var_pos_file
9. Pair energy
lookup table generation
9.1 Overview
Figure 9.1.1 - Pairwise
decomposition of energies
9.1.1 Rotamer
complexity, disk and memory usage
Figure 9.1.1.1 - Rotamer complexity
and memory usage
9.2 Inputfile
options
9.2.1 LOOKUP_TABLE_DIRECTORY
9.2.2 PRECALCULATION_LEVEL
9.2.3 JOBTYPE LOOKUP_TABLE_SLAVE
9.3 Brief description of
lookup table generation
9.4 Parallelization of lookup
table generation
9.4.1 Brief
description of parallelized lookup table generation
9.5 Description of
lookup table
generation code
9.5.1
Description of LOOKUP_ENERGY datastructures
Figure 9.5.1.1
- Schematic of EGAD datastructures
9.5.2
Discussion and rationale
for the implemented strategy, thoughts on improvement
9.5.3
Description of lookup table
generation functions
9.5.4
Description of parallel
lookup table generation functions
10. Rotamer
optimization jobs
10.1 Overview
10.2 Algorithms
10.2.1 JOBTYPE faster
10.2.2
Monte carlo simulated
annealing (JOBTYPE mc)
Figure
10.2.2.1 - Monte Carlo simulated annealing
10.2.3
Genetic algorithm (JOBTYPE ga)
Figure
10.2.3.1 - Genetic algorithms
10.2.4
Heuristic quench
minimization (HQM)
10.2.5
JOBTYPE
mc_ga or JOBTYPE sa_ga
10.2.6
Dead-end elimination (JOBTYPE dee)
10.2.7
Self-consistent mean field
optimization (JOBTYPE scmf)
Figure 10.2.7.1
- Self-consistent mean field
optimization
10.2.8 OUTPUT_PREFIX.xxx.out format
10.3
User-definable options
10.3.1
Simulation
temperatures
10.3.2
RUNTIME
10.3.3
Logfile options
10.4 Premature
program termination
10.5 Strengths and
weaknesses of different rotamer
optimization methods
10.6 Implementation details
10.6.1 Data structures
and general-use functions
10.6.2 Genetic
algorithm (GA)
10.6.2.1 Control functions for genetic
algorithm
10.6.2.2
Actual genetic algorithm: GA.cpp:
evolve_sidechains
10.6.3
Monte carlo
simulated annealing (MC)
10.6.3.1 Control function MC.cpp:
MC_rotamer_control
10.6.3.2 Actual simulated annealing: MC.cpp:
MC_rotamers
10.6.4 Heuristic
quench minimization (HQM): HQM_rotamers.cpp:
HQM_rotamers
10.6.5
Fast and accurate
side-chain topology and energy refinement (FASTER)
10.6.6
Dead-end elimination (DEE)
10.6.7
Self-consistent mean field
optimization (SCMF)
10.6.8
Simulation temperature
11.
Prediction of pKa's - JOBTYPE
pK
Figure 11.1 - Thermodynamic cycle for
pK calculations
Figure 11.2 - pK-shifts are coupled to
electrostatic energies
Figure 11.3 -
Quality of EGAD pKa predictions
12. Ligands
12.1 Overview
12.1.1 Multiple ligand
choices
12.2 Ligand binding
protein
design examples
12.2.1 Redesign of
maltose binding protein (MBP) for maltose
12.2.2
Redesign of maltose binding
protein for an alternative ligand
12.2.3
Drug design - design of an N-substituted peptoid for binding src SH3
Figure 12.2.3 - Drug design figures
12.3 Ligand
hydrogen-bonding satisfaction: UNSAT_HBOND_REFSTATE
12.4 Ligand implementation
13.
Parallelization of EGAD jobs
13.1 Overview
13.2 Requirements
13.3 Inputfile options
13.4 Launching a parallel
job
13.5 CPU_FILE
13.6 Implementation of
parallelization for rotamer
calculations
14. Scanning
mutagenesis
14.1 Overview
Figure 14.1.1 - Mutant stability
prediction thermodynamic cycle
Figure 14.1.2 - Summary of mutant
stability prediction
Figure 14.1.3 - Examples of
mutant stability predictions
14.2 alanine_scan and saturation_mutagenesis
15. Batch jobs
16.
Protein-protein complexes
16.1 Overview
Figure 16.1.1 - Mutant affinity
prediction thermodynamic cycle
Figure 16.1.2 - Summary of mutant
affinity prediction
Figure 16.1.3 - Examples of
mutant affinity predictions
16.2 JOBTYPE
separate_complex or JOBTYPE
split_complex
16.3 JOBTYPE
complex_formation_energy
16.4 JOBTYPE
interface_alanine_scan and interface_saturation_mutagenesis
16.5 Implementation
17.
Sequence restraints on design
17.1 MAX_MUTATIONS
modifier
17.2
User-defined sequence restraint
functions
18. Analyses and
optimizations that require
consideration of multiple structures/states
18.1 Target and competitor
structures
18.2 Defining positions to
optimize and/or move
18.3 MULTISTATE_JOBTYPE
18.3.1 MULTISTATE_JOBTYPE scan
18.3.2 MULTISTATE_JOBTYPE ga
18.3.3 MULTISTATE_JOBTYPE quench
18.3.4
Initialization of optimization
18.4 Slave
parameters, MASTER_OPTIMIZATION_TIME, and early
termination
18.5 Objective functions
for multistate design
Figure
18.5.1 - Energy spectrum for multistate protein design
18.5.1 Custom
multistate objective functions
18.6 Examples of multstate
design
18.6.1 Designing
binding specificity: change glutamine-binding protein to bind glutamate
Figure
18.6.1.1 - Design of glutamate-binding specificity in glutamine-binding
protein
18.6.2 Designing
conformational specificity: lock maltose-binding protein in the closed
state
18.6.3
Design of a specific
protein-protein interface: src SH3 domain peptide-binding site
Figure
18.6.3.1 - Design of src SH3 binding to a target peptide
19. Building the
backbone with ideal geometry
19.1 Overview
19.2 Example -
grafting and stitching of
backbone fragments
20. Dihedral
angle minimization - JOBTYPE minimize
20.1 Overview
20.2 Moving the
backbone
20.3 Restricting moving
positions
20.4 Minimization exit
Appendices:
A.1
Output
pdb file format
A.1.1 ENERGY entries
A.1.2 ROTAMER
entries
A.1.3 SEQUENCE
entry
A.1.4 CHARGE,
HBOND, CLASH, LENGTH,
and CENTROID entries
A.1.5 ATOM
entries
A.1.6 Output pdb
file options
A.1.7 Implementation and
use
A.2 Energy
function file formats
A.2.1
Forcefield file
A.2.1.1 General format
A.2.1.2 Atomtype line
format
A.2.1.3
Constant line and file
line formats
A.2.1.4
Comments
A.2.1.5
Description of the
current file and tips for modification
A.2.2 Torsion parameter
file
A.2.2.1 General format
A.2.2.2
Torsion parameter format
A.2.2.3
Comments
A.2.2.4
Description of the
current file and tips for modification
A.2.3 Resparam
files
A.2.3.1 General format
A.2.3.2 Residuetype
header formats
A.2.3.3 Atomline format
A.2.3.4
Comments
A.2.3.5
Description of the
current file
A.2.4 Rotamer
library file
A.2.4.1 General format
A.2.4.2
Rotamer headerline - Old
version
A.2.4.3
Rotamer headerline - New
version
A.2.4.4
Rotamerline format
A.2.4.5
Position-specific rotamers
A.2.4.6
Comments
A.2.4.7
Description of current
file and tips for modification
A.2.5 Ligand
parameter files
A.2.5.1 General format
A.2.5.2 Resparam entry
section
A.2.5.2.1 Generation of resparam section
from a ligand pdb file
A.2.5.3 Ligamer section
A.2.5.3.1 Automatic
generation of ligamers
A.2.5.3.2
Ligamer line format
A.2.5.4
Base coordinate section
A.2.5.5
Ligamers defined by
coordinates
References
Examples directory
Source code directory
download EGAD User's Manual (pdf)
download Navin Pokala's thesis (pdf)
download the EGAD
program
with example files
installation instructions
back to
EGAD homepage